Patients with systolic heart failure (HF) are thought to be at increased risk for stroke and thromboembolic complications, potentially contributing to hospital readmissions and mortality.1–3 Some literature reports demonstrate that patients with HF are hypercoaguable because of increased platelet activation and elevated coagulation markers such as d-dimer, β-thromboglobulin, and thrombin-antithrombin III complexes that could lead to thrombus formation.3,4 Thrombi are thought to arise from diseased left ventricular (LV) or left atrial endocardia secondary to blood stasis. The incidence of thromboembolism in congestive HF patients has been reported to be 1% to 3% per year.5
The etiology of HF and/or presence of comorbid conditions often determine guideline recommendations for or against warfarin therapy. The etiology of systolic LV dysfunction is thought to be approximately 70% ischemic and 30% nonischemic.6 Currently, the 2008 American College of Chest Physicians guidelines regarding the use of anticoagulants recommends against the use of oral vitamin K antagonists in patients with nonischemic HF unless another compelling indication is present.6 The American College of Cardiology/American Heart Association 2009 guidelines for HF state that warfarin is most justified in patients with HF who have other indications for anticoagulation, such as a previous embolic event or atrial fibrillation.7 The 2009 American College of Cardiology/American Heart Association ST-elevation myocardial infarction (STEMI) guidelines state it is reasonable (but not necessarily recommended) to prescribe warfarin to post-STEMI patients with LV dysfunction and extensive wall motion abnormalities (Class IIa, level of evidence A). A weaker class IIb (level of evidence C) recommendation is to consider warfarin in post-STEMI patients with severe LV dysfunction, with or without congestive HF.8 The 2008 American College of Chest Physicians guidelines also address warfarin therapy in high-risk post-MI patients with large anterior MI, those with significant HF, those with intracardiac thrombus, those with atrial fibrillation, and those with history of an embolic event. A Grade 2A recommendation (treatment is reasonable because the risks might outweigh the benefits) states to initiate moderate-intensity (international normalized ratio [INR], 2.0–3.0) oral anticoagulation for at least 3 months post-MI.6
THROMBOEMBOLIC EVENTS: CLINICAL TRIALS DATA
As mentioned above, patients with LV dysfunction may be in a hypercoaguable state in addition to having a potential nidus for thrombus formation due to poor blood movement inside the ventricular chamber. The rationale for warfarin therapy in patients with systolic HF is therefore to decrease the risk of arterial thromboembolic events such as stroke. Table 1 summarizes the results of studies, which analyzed thromboembolic events as study outcomes in patients with LV systolic dysfunction.9–14 Four studies, Studies of Left Ventricular Dysfunction (SOLVD), the Veterans Affairs Cooperative Study Vasodilator-Heart Failure Trials (V-HeFT I and II), and Survival and Ventricular Enlargement (SAVE), are post hoc analyses of larger randomized controlled trials9–11 and 3 studies, The Warfarin/Aspirin Study in Heart Failure (WASH), Warfarin and Antiplatelet Therapy in Heart Failure (WATCH), and Heart failure Long-term Antithrombotic Study (HELAS), are smaller randomized controlled trials.12–14
The SOLVD subgroup analysis looked at 861 warfarin users out of 6513 patients in the original SOLVD studies.9 Patients were included in the original studies if they had LV systolic dysfunction (67% ischemic, 32% nonischemic) defined as a left ventricular ejection fraction (LVEF) ≤35%. The original studies were designed to evaluate the efficacy and safety of enalapril for the treatment of patients with LV dysfunction. Patients were on warfarin at the investigator's discretion. The subgroup analysis of warfarin users showed a reduction in all-cause mortality (hazard ratio [HR], 0.76; 95% confidence interval [CI] [0.65–0.89]), death or hospital admission for HF (HR, 0.82; 95% CI [0.72–0.93]), and cardiovascular deaths in patients with warfarin therapy (HR, 0.72; 95% CI [0.61–0.86]). There was no reduction in fatal noncardiac vascular events, including deaths due to stroke, pulmonary embolism (PE), or other vascular causes. More patients in the warfarin group had atrial fibrillation (19.3% vs. 4.5%) and cerebrovascular disease (13.8% vs. 5.5%) compared with the nonwarfarin group. This analysis demonstrated that warfarin may be helpful in patients with LV dysfunction, but many of these patients had other indications for warfarin, and therefore these results cannot necessarily be extrapolated to the HF patient with no other indication for oral anticoagulation.
The V-HeFT studies were designed to look at the efficacy and safety of hydralazine and isosorbide dinitrate (V-HeFT I) and enalapril (V-HeFT II) in patients with systolic heart failure. Warfarin use in these patients was at the discretion of the investigator. In V-HeFT I, the warfarin subgroup analysis (n = 642) looked at the incidence of stroke, peripheral embolus, and PE. A similar event rate was seen in warfarin users (2.9/100 patient-years) versus nonusers (2.7/100 patient-years) (P = not significant). As expected, there were higher incidences of stroke and thromboembolism in patients with atrial fibrillation and prosthetic valves not on anticoagulation.
The V-HeFT II trial looked at 804 men with HF symptoms (53% had coronary artery disease) and an EF <45%.10 This warfarin subgroup analysis looked at the same outcome events as V-HeFT I. A higher event rate was seen in warfarin users (4.9/100 patient-years) versus nonwarfarin users (2.1/100 patient-years) (P = 0.01). More patients on anticoagulants had a stroke or thromboembolism with prosthetic valves (8.8% vs. 5.1%) compared with no anticoagulants. However, the INR for patients on warfarin was not recorded, the number of patients within a therapeutic INR was not recorded, and the duration of time within a therapeutic INR was not recorded. Similar to V-HeFT I, there were more thromboembolic events in patients with atrial fibrillation not receiving anticoagulation compared with those with atrial fibrillation receiving oral anticoagulation (5.6% vs. 3.8%; P = 0.45). Both V-HeFT I and II analyses showed no significant decrease in the number of thromboembolic events in patients treated with oral anticoagulation, regardless of accepted indications for anticoagulation such as atrial fibrillation and prosthetic heart valves.
The SAVE trial looked at the efficacy and safety of captopril in patients with MI and LV dysfunction (EF <40%, measured 2–16 days post-MI).11 A post hoc analysis studied the relationship between LVEF and the incidence of stroke in patients with LV dysfunction, but without symptoms of HF, to determine if LVEF is a risk factor for stroke. There was an 18% increase in the risk of stroke for every absolute decrease of 5% in LVEF from baseline (average EF, 31%). The multivariate analysis looking at the protective effects of oral anticoagulation showed a relative risk (RR) of 0.17 (CI, 0.09–0.29; P < 0.001) for LVEF ≤28%; RR, 0.14 (CI, 0.06–0.28; P < 0.001) for LVEF 29% to 35%; and RR, 0.23 (CI, 0.12–0.47; P < 0.001) for LVEF >35%. In other words, warfarin showed an 83%, 86%, and 77% reduction in risk of stroke for patients with LVEF of ≤28%, 29% to 35%, and >35%, respectively. Most of the strokes (96%) were of ischemic origin. Very few patients received oral anticoagulants in either group (n =2231). In patients who developed stroke, 38% were on anticoagulants; of those who did not develop stroke, 28% were on anticoagulants. Also, more patients who experienced a stroke had atrial fibrillation or atrial flutter versus those who did not experience a stroke (16% vs. 10%). It is unknown how many of these patients with atrial fibrillation or atrial flutter were on anticoagulation. Like the V-HeFT II trial, there were no records of INRs or how many patients achieved therapeutic INR values and for how long. This analysis demonstrated that as LVEF decreases the risk of stroke increases.
WASH was a pilot study that included patients with LVEF <35% (majority had ischemic HF).12 The purpose of this study was to see if a larger outcome study of warfarin versus aspirin in patients with HF would be feasible. Following 2 comparator arms were analyzed in this study: (1) aspirin 300 mg daily, (2) warfarin with a target INR of 2.5 (range, 2.0–3.0), or (3) placebo. The primary outcome of this study was to compare the effects of no antithrombotic therapy, aspirin, or warfarin on the composite of death, nonfatal MI, or nonfatal stroke. This study was stopped early due to slow recruitment and published results showed that there was no significant difference in the composite end point among the 3 arms.12 This study was not blinded and had insufficient power to detect a difference if there was one. Even though the mean INR throughout the study was 2.3 (goal, 2.0–3.0), it was not mentioned how many patients achieved that value or for how long.
The WATCH Trial13 randomized patients (n = 1587) with symptomatic HF with an EF and in sinus rhythm 3 months to 3 different treatment arms as follows: aspirin 162 mg or clopidogrel 75 mg provided in a double-blind, double-dummy manner, or open-labeled warfarin titrated to an INR goal of 2.5 to 3.0. The primary end point was a composite of all-cause mortality, nonfatal MI, and nonfatal stroke. After 18 months of slow enrollment, the study was terminated early and results were published. The statistical analyses were adjusted for low sample size and an intent-to-treat log-rank test was used to compare the effects of treatment on total mortality. For the composite primary outcome, there was no difference between aspirin and warfarin with an unadjusted HR of 0.98 (95% CI, 0.86–1.12; P = 0.77). The HR for warfarin versus clopidogrel was 0.89 (95% CI, 0.68–1.16; P = 0.39). The authors concluded that the results of their study do not support the hypothesis that warfarin is superior to aspirin in preventing major cardiovascular outcomes in patients with LV dysfunction in sinus rhythm. The results of this study should be interpreted with caution secondary to early termination and the study being under powered to detect differences in outcomes.
The HELAS study was a multicentered, randomized, double-blind, placebo-controlled trial in 197 patients with LVEF <35%.14 Patients were separated into an ischemic HF group who received either aspirin 325 mg daily or warfarin with an INR target of 2.5 (range, 2.0–3.0) and the second group was comprised of idiopathic dilated cardiomyopathy patients randomized to receive either warfarin with a target INR of 2.5 (range, 2.0–3.0) or placebo. The primary end point was a composite of nonfatal stroke, peripheral thrombus or PE, myocardial (re)infarction, rehospitalization, and exacerbation of HF or death from any cause. For the ischemic HF arm, the primary end point was similar between the aspirin and warfarin groups: 14% versus 13%, respectively. For the idiopathic dilated cardiomyopathy patients, the primary end point occurred in 10% for the placebo group versus 6% for the warfarin group (P value not reported). Due to low patient enrollment, differences in efficacy between treatment groups could not be convincingly determined. Even though there was a goal INR of 2.0 to 3.0 stated, it was not reported how many patients reached goal INR and it is unknown how long patients stayed in that goal INR range.
BLEEDING EVENTS IN CLINICAL TRIALS
Bleeding complications during the use of warfarin therapy can be significant and life-threatening. None of the post hoc analyses reported bleeding rates in patients treated with warfarin. In the prospective trials, these adverse events were reported.
The WASH trial reported 5 major hemorrhagic events; 4 of them occurred in patients treated with warfarin and the other occurred in a patient receiving aspirin.12 There were no reported major bleeding events in patients who received no antithrombotic therapy. All of the reported major bleeding events were gastrointestinal bleeds. Only 1 of the 4 major bleeds had an INR recorded which was subtherapeutic at 1.3. The other 3 major bleeds did not have INRs recorded during the event. Minor bleeding events were also higher in the aspirin and warfarin groups compared with no antithrombotic therapy (13%, 17%, and 5%, respectively; P = 0.033). Major and minor bleeding was not defined.
The WATCH study reported both major and minor bleeding episodes.13 There were more episodes of major bleeding in the warfarin group compared with the clopidogrel group (P < 0.01) but not the aspirin group (P = 0.22). There were 7 central nervous system (CNS) bleeds in 6 patients in the warfarin group, 3 CNS bleeds in the aspirin group, and 1 CNS bleed in the clopidogrel group. There were more minor bleeds reported in the warfarin group than in the clopidogrel group (P = 0.025) and a similar trend in the warfarin versus aspirin group (P = 0.054).
The HELAS study reported major hemorrhagic events; 4 occurred in the ischemic cardiomyopathy patients treated with warfarin and 3 occurred in the dilated cardiomyopathy patients treated with warfarin.14 This averaged to a rate of 4.6 major hemorrhages per 100 patient years in patients treated with warfarin. The authors reported that these events were likely secondary to over-anticoagulation (no INR values were reported). None of these events led to fatalities. There were no major hemorrhages reported in the aspirin or placebo groups.
Warfarin therapy is not without risk as bleeding complications with oral anticoagulation are common and can be life-threatening. These events occur at varying rates and have been reported between 10 to 17 events per 100 patient-years for all bleeding complications and from 2 to 5 per 100 patient-years for major bleeding complications.15–22 Patients with HF receiving anticoagulation have additional risks including increased age, fluctuating liver function secondary to congestion, and drug interactions that can complicate therapy with a narrow therapeutic index drug such as warfarin.15–22 It is also important to consider other barriers to effective and safe warfarin therapy including the frequency of visits to a health care practitioner for INR checks and consistency in diet. Some patients may find it difficult to see a provider as often as necessary for titration of warfarin dosage and others may be unable to abide by dietary restrictions, secondary to noncompliance or variation in appetite secondary to their disease state.
Studies completed over 50 years ago reported benefits from warfarin anticoagulation in certain populations of HF patients, the majority of whom had atrial fibrillation and rheumatic heart disease.23–26 However, when considering long-term oral anticoagulation in a broader HF population, a review of the literature does not provide clear evidence to anticoagulate for the reduction of stroke and thromboembolic events.
The post hoc analyses of large HF trials (SOLVD, V-HeFT I/II, and SAVE) reported conflicting results.9–11 It should be noted that all of these trials were retrospective, observational analyses in which patient populations differed (ischemic vs. nonischemic, symptomatic vs. asymptomatic, comorbidities such as atrial fibrillation), and the use of warfarin was at the discretion of the investigator. Target INRs, the percentage of time adequately anticoagulated, and INRs at the time of thromboembolic events were not recorded.
The prospective trials that aimed to determine the effectiveness of anticoagulation in HF patients for the prevention of thromboembolic events all fell short in their ability to recruit enough patients.12–14 There were numerous reasons why these trials fell short. The WASH trial did not fair well with recruitment possibly because there was hesitation to withhold antithrombotic treatment from HF patients. For this same reason funding agencies were reluctant to participate so funding might have been an issue in this study.12 The WATCH trial was stopped short because their funding was cut off. This study was funded by the Veteran's Affairs with a modest budget to begin with. It is thought the inclusion criteria was too strict because of a limitation of New York Heart Association class II patients, having a 30% EF threshold, and requiring the loop diuretics be at certain dose.13 There were also high rates of turnover of study site coordinators, with very little training for replacement coordinators.13 Lastly, since this trial was an open-label study for warfarin, there were concerns brought up from the investigators, practitioners, and patients at various study sites about the use of study drugs. Some practitioners and investigators did not want some patients to use 1 of the study drugs or wanted some patients to use a certain drug over the others. Some patients did not want to take warfarin because of the inconvenience and risks described to them in the consent forms.13 The HELAS study showed unexpectedly slower recruitment and was forced to terminate enrollment early; this study is therefore deemed to be a pilot study.14 The trials may have published results, but these data should be cautiously interpreted secondary to inadequate power to detect a difference among treatment groups.
The Warfarin vs. Aspirin in Patients with Reduced Cardiac Ejection Fraction (WARCEF) trial is a randomized, double-blind clinical trial currently underway looking at patients in normal sinus rhythm and having an LVEF ≤35%.27 Patients are being randomized to warfarin plus placebo or aspirin plus placebo to determine a difference in the composite end point of death or stroke (ischemic and hemorrhagic). The trial will also compare the prevention of stroke alone. This trial might answer more questions regarding the clinically significant issue of routinely anticoagulating patients with HF and LV dysfunction.
Warfarin is sometimes prescribed to patients with HF and LV dysfunction to prevent stroke and thromboembolic events. However, published studies suggest that warfarin may not be beneficial in preventing these events in patients without other indications for anticoagulation, such as atrial fibrillation, a suspected blood clot, or a history of blood clots. The rate of thromboembolic events is relatively small in HF and the risk of bleeding is evident. With the inconsistent data currently reported in the literature regarding the efficacy of oral anticoagulation in these patients, clinicians need to weigh the risks and benefits of anticoagulation in each patient based on the etiology of HF (ischemic vs. nonischemic), comorbidities, compliance with medication adherence and clinic appointment attendance, fall risk, concurrent medications, and the patients ability and desire for close monitoring of a narrow therapeutic window medication. In the contemporary practice of evidence-based medicine, the current evidence does not support the routine use of oral anticoagulation in patients with LV dysfunction and no other indication for anticoagulation.
1.Cleland JG. Anticoagulation and antiplatelet therapy in heart failure. Curr Opin Cardiol
2.Loh E, Sutton MS, Wun CC, et al. Ventricular dysfunction and the risk of stroke after myocardial infarction. N Engl J Med
3.Lip GY, Gibbs CR. Does heart failure confer a hypercoagulable state? Virchow's triad revisited J Am Coll Cardiol
4.Jafri SM. Hypercoagulability in heart failure. Semin Thromb Hemost
5.Hunt SA, Abraham WT, Chin MH, et al. 2009 Focused update incorporated into the ACC/AHA 2005 guidelines of the diagnosis and management of heart failure in adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines: developed in collaboration with the International Society for Heart and Lung Transplantation. Circulation
6.Becker RC, Meade TW, Berger PB, et al. The primary and secondary prevention of coronary artery disease: American College of Chest Physicians Evidence-Based Cinical Practice Guidelines (8th Edition). Chest
7.Hunt SA, Abraham WT, Chin MH, et al. 2009 focused update incorporated into the ACC/AHA 2005 guidelines of the diagnosis and management of heart failure in adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines: developed in collaboration with the International Society for Heart and Lung Transplantation. Circulation
8.Kushner FG, Hand M, Smith SC, et al. 2009 focused updates: ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction (updating the 2004 guidelines and 2007 focused update) and ACC/AHA/SCAI guidelines on Percutaneous Coronary Intervention (updating the 2005 guidelines and 2007 focused update). Circulation
9.Al-Khadra AS, Salem DN, Rand WM, et al. Warfarin anticoagulation and survival: a cohort analysis from the Studies of Left Ventricular Dysfunction. J Am Coll Cardiol
10.Dunkman WB, Johnson GR, Carson PE, et al. The V-HeFT VA Cooperative Studies Group. Incidence of thromboembolic events in congestive heart failure. Circulation
11.Loh E, St John Sutton M, Wun CC, et al. Ventricular dysfunction and the risk of stroke after myocardial infarction. N Engl J Med
12.Cleland JG, Findlay I, Jafri S, et al. The Warfarin/Aspirin Study in Heart Failure (WASH): a randomized trial comparing antithrombotic strategies for patients with heart failure. Am Heart J
13.Massie BM, Collins JF, Ammon SE, et al. Randomization trial of warfarin, aspirin, and clopidogrel in patients with chronic heart failure: the Warfarin and Antiplatelet Therapy in Chronic Heart Failure (WATCH) Trial. Circulation
14.Cokkinos DV, Haralabopoulos GC, Kostis JB, et al. Efficacy of antithrombotic therapy in chronic heart failure: the HELAS study. Eur J Heart Fail
15.Fihn SD, McDonell M, Martin D, et al. Warfarin Optimized Outpatient Follow-up Study Group. Risk factors for complications of chronic anticoagulation: a multicenter study. Ann Intern Med
16.Landefeld CS, Beyth RJ. Anticoagulant-related bleeding: clinical epidemiology, prediction and prevention. Am J Med
17.Levine MN, Raskob G, Landefeld SC, et al. Hemorrhagic complications of anticoagulant treatment. Chest
. 1995;108(suppl 4):276–289.
18.Van der Meer FJ, Rosendaal FR, Vandenbroucke W, et al. Bleeding complications in oral anticoagulant therapy. Arch Intern Med
19.Van der Meer FJ, Rosendaal FR, Vandenbroucke W, et al. Assessment of a bleeding risk index in two cohorts of patients treated with oral anticoagulants. Thromb Haemost
20.Parlaret G, Leali N, Coccheri S, et al. Bleeding complications of oral anticoagulant treatment: an inception-cohort, prospective collaborative study (ISCOAT). Lancet
21.Gitter MJ, Jaeger TM, Petterson TM, et al. Bleeding and thromboembolism during anticoagulant therapy: a population-based study in Rochester, Minnesota. Mayo Clin Proc
22.The Stroke Prevention in Atrial Fibrillation Investigators. Bleeding during antithrombotic therapy in patients with atrial fibrillation. Arch Intern Med
23.Anderson GM, Hull E. The effect of dicumarol upon the mortality and incidence of thromboembolitic complications in congestive heart failure. Am Heart J
24.Griffith GC, Strangnell R, Levinson DC, et al. A study of the beneficial effects of anticoagulant therapy in congestive heart failure. Ann Intern Med
25.Harvey WP, Finch CA. Dicumarol prophylaxis of thromboembolic disease in congestive heart failure. N Engl J Med
26.Wishart JH, Chapman CB. Dicumarol therapy in congestive heart failure. N Engl J Med
27.Pullicino P, Thompson JL, Barton B, et al. Warfarin versus aspirin in patients with reduced cardiac ejection fraction (WARCEF): rationale, objectives, and design. J Card Fail